AI Regulation & Policy

Today's Stories

1. 1

   Walmart CEO John Furner: no AI personalized pricing

   In a Sept. 25 letter, Walmart CEO John Furner said the retailer will not use AI, customers' income, shopping history or other personal information to charge different shoppers different prices. The pledge rules out charging more based on a customer's identity, time of day, income, shopping history, urgency, or what Walmart thinks the customer could afford, Furner said.

   Furner said the policy also covers Sparky, Walmart's AI shopping assistant, and digital shelf labels, which are meant to match shelf prices to checkout prices; whether the commitments hold as the technology changes is the test.

2. 2

   S&P: governance to decide AI winners among insurers

   S&P Global Ratings surveyed 121 rated re/insurance entities (about 38% of the assets it rates in the sector) and said governance maturity, not adoption speed, will decide which insurers benefit; it warned weaknesses could hurt credit quality. The ratings agency said no rating actions tied to these systems have occurred yet, but that differences in readiness and data capabilities may increasingly influence competitive advantage and creditworthiness.

   EIP CEO Ross Sinclair said the key question is whether AI only gathers information or actually decides claims; watch whether AI budget expansion over the next three years changes that line.

3. 3

   Oracle sends "force majeure" notice on $165 billion data center

   Oracle sent a "force majeure" notice to Blue Owl Capital on its $165 billion data center project, seeking to pause rent payments if delays push past 2028. Oracle says its 2.45-gigawatt New Mexico facility remains on track for 2028 but flagged the state's regulatory environment. The notice shows that even a large, well-capitalized builder sees real legal and regulatory risk in finishing data centers on schedule. It is up to Blue Owl Capital to honor or challenge the request in court.

   The outcome hinges on whether Blue Owl accepts the pause or contests it, and on New Mexico's regulatory hurdles. The project is still guided for completion in 2028.

4. 4

   Google, OpenAI, Anthropic make AI safety move

   Google, OpenAI, and Anthropic announced a joint move on AI safety. Rather than waiting for action from Washington, the three companies are building something of their own. The three biggest names in AI acting together on safety, without waiting for US regulators, suggests they intend to shape safety standards themselves.

   Their approach will be tested against whatever Washington eventually does. The key question is whether voluntary industry action preempts or complements regulation.

5. 5

   Hinton: AI may want to get rid of people

   Hinton told the Atlantic that a smart AI given a goal will derive its own subgoals, and even without a bad actor, those subgoals may make it want to get rid of people. If Hinton is right, the danger can arise from ordinary goal-seeking rather than only from someone deliberately assigning harmful goals, which is why he says the government needs independent evaluators to test models.

   Hinton also argued that slowing frontier development is better than nothing but not good enough, and suggested regulation should steer AI like a car's steering wheel rather than act as brakes.

6. 6

   US, China set AI incident channel after three-day summit

   After a three-day summit in Washington, both sides agreed to a communication mechanism for AI-related incidents and an AI dialogue in November, plus a military crisis-communications memorandum. The summit produced no major breakthroughs, but experts say the working groups and military-to-military channel could keep problems from escalating, according to the body.

   The steps are light on details and hinge on whether working-level engagement continues; watch the November AI dialogue and APEC summit in Shenzhen.